火力发电厂为工业生产和居民生活提供最基本的电力需求保证，我国目前50%的原煤直接用于火力发电厂进行燃煤发电，在此过程中需要将原煤进行粉碎、研磨等工序加工成煤粉，在送入燃烧锅炉中使其充分燃烧。大型电厂的研磨工艺多数采用立式磨煤系统，为了除去原煤中水分对煤粉的影响，会向磨煤系统预先通入热空气，煤粉粒径小（20~50μm）且磨煤机为干燥气氛（温度在60℃以上）容易发生煤粉自燃，因此需要可靠的预报技术提前预警。 本文通过参考相关文献和实地调研，选取五种不同变质程度煤种，粉碎研磨到与电厂燃用煤粉粒径相同的范围，采用煤自燃程序升温装置对煤粉自燃影响因素及预报指标气体进行了实验研究，为保证预报参数指标的有效性，实验过程中模拟电厂磨煤系统粉磨环境预先通入干燥空气加热升温箱，使升温箱起始温度在60℃左右开始实验。在实验和理论相结合的研究方法下，得出电厂不同煤粉自燃预警指标气体和参数值。为了减少实验的偶然性并给现场提供更多工况下煤粉自燃的预报指标，改变实验参数并对煤粉自燃预报指标作进一步研究；通过改变煤粉粒径、实验供风量和升温速率等实验条件，测试煤粉自燃氧化指标种类及含量与温度的变化情况。 由于煤种不同，选取的预报指标也不同，在实际生产中为防止同一煤种的预报指标不准确甚至互相冲突的影响，采用灰色关联分析法，将实验测试的指标气体及参数值进行可靠度评价，即通过计算各个指标值与煤粉自燃可能性的关联程度大小，评价指标气体的可靠度，选取与煤种关联度较高的指标作为相应煤粉的自燃预报指标。 电厂制粉系统存在煤粉自燃危险，只有及时准确的预测预报才能减少煤粉自燃对生产安全的影响，因此研究得出的煤粉自燃预警指标对电厂煤粉自燃预防有一定现实意义。

Thermal power plants for industrial production and residential life to provide the most basic demand for electricity to ensure that China's current 50% of the coal directly used in thermal power plants for coal-fired power generation, in the process of raw coal need to crush, grinding and other processes into coal Powder, into the combustion boiler to make it fully burned. Large-scale power plant grinding process using the majority of vertical coal mill system, in order to remove the raw coal water on the impact of coal, coal mill system will be pre-access to hot air, coal particle size is small (20 ~ 50μm) and the mill Dry atmosphere (temperature above 60 ℃) prone to spontaneous combustion of coal, so the need for reliable forecast technology ahead of warning. Based on the relevant literature and field investigation, this paper chooses five kinds of coal samples with different metamorphic degrees by means of theoretical analysis and experimental verification. The coal is pulverized and pulverized into the same range of pulverized coal with the same power plant. In order to ensure the effectiveness of the forecasting parameter index, the coal preparation system of the power plant is preheated by the hot air drying and the heating chamber, and then the coal spontaneous combustion process is carried out for each coal type. In this paper, Temperature test. Under the combination of experiment and theory, the gas and parameter values of coal spontaneous combustion warning index are obtained. In order to reduce the accident of the experiment and to provide more indicators of coal spontaneous combustion in the field, the experimental parameters were changed and the spontaneous combustion forecast of pulverized coal was further studied. By changing the coal particle size, air flow and heating rate and other parameters of the conditions, test the spontaneous combustion of coal powder gas species and concentration changes. Because of the different coal types, the selected forecast indexes are different. In the actual production, the factors such as the inaccurate and even the conflicting index of the same coal are prevented, and the reliability of the index gas and the parameter value of the experiment is evaluated by gray correlation analysis , That is, by calculating the correlation degree between the index gas and the coal spontaneous combustion program temperature experiment, the reliability of the index gas is evaluated, and the first three kinds of gas with the highest correlation degree are selected as the spontaneous combustion forecasting gas of the corresponding pulverized coal. There is a danger of spontaneous combustion of pulverized coal in power plant pulverizing system. Only timely and accurate forecasting can reduce the influence of spontaneous combustion of pulverized coal on production safety. Therefore, it is pointed out that the spontaneous combustion early warning index of pulverized coal has certain guiding significance for coal spontaneous combustion prevention.